Population Genetics of Melampsora larici-populina — Microsatellite Analysis

An R Markdown workflow for population genetic analysis of poplar rust (M. larici-populina) isolates using microsatellite markers, including lineage definition, clustering, reproductive mode inference, spatial mapping, and regression.

This code workflow is part of the study titled:

Long-lasting coexistence of multiple asexual lineages alongside their sexual counterparts in a fungal plant pathogen

Ammar Abdalrahem, Axelle Andrieux, Ronan Becheler, Sébastien Duplessis, Pascal Frey, Benoit Marcais, Kadiatou Schiffer-Forsyth, Solenn Stoeckel, Fabien Halkett

---

What the Code Does

1. Data loading — Reads isolate metadata and microsatellite genotypes from Table_data.tsv
2. MLG/MLL definition — Defines multilocus genotypes and multilocus lineages using distance-based thresholding (poppr)
3. Clustering — Assigns isolates to genetic clusters via K-means + DAPC (adegenet); removes uncertain assignments (<80% posterior probability)
4. Reproductive mode inference — Classifies isolates as asexual or sexual using two approaches:
   • Cluster approach: negative Fis → asexual (indirect indication)
   • Resampling approach: MLL persisting across ≥ 2 years → asexual
   • Final label: asexual if flagged by either approach
5. Population genetic indices — Calculates N, G/N, Ar, Ho, Hs, Fis, rbarD per group (hierfstat, poppr)
6. Phylogenetic tree — Builds bootstrapped UPGMA/NJ tree annotated by lineage and reproductive mode (ggtree)
7. Spatial analysis — Maps sexual vs. asexual proportions across France (mapdata, mapplots)
8. Regression — GLM and GLMM testing effect of latitude, longitude, and year on reproduction mode (lme4)

---

Dependencies

R version ≥ 4.3 and RStudio (required for automatic path detection) are recommended.

Install all CRAN packages:

install.packages(c(
  "knitr", "ggplot2", "readxl", "tidyverse", "genepop", "hierfstat",
  "mapdata", "mapplots", "data.table", "colorspace", "adegenet", "poppr",
  "pegas", "ape", "cowplot", "ade4", "remotes", "ggtreeExtra", "viridis",
  "factoextra", "openxlsx", "ggrepel", "RClone", "ggsci", "scales",
  "lme4", "dplyr", "reshape2", "rstudioapi"
))

Install the Bioconductor package ggtree:

if (!requireNamespace("BiocManager", quietly = TRUE)) install.packages("BiocManager")
BiocManager::install("ggtree")

---

Input Data

File	Description
Table_data.tsv	Isolate metadata + microsatellite genotypes (must be in the same folder as the .Rmd file)

• Microsatellite locus columns must contain Mlp in the name
• Missing genotype values must be coded as 999

---

How to Run

In RStudio (recommended):

1. Place data_analysis_mlp.Rmd and Table_data.tsv in the same folder
2. Open the .Rmd file in RStudio
3. Install dependencies (see above)
4. Click Knit → renders a full HTML report

From the R console:

setwd("path/to/your/folder")
rmarkdown::render("data_analysis_mlp.Rmd")

---

Outputs

File	Description
data_analysis_mlp.html	Full rendered report
new_genotype_data.csv	Final genotype table with reproduction labels
filtered_mll_years.csv	MLLs recurring across multiple years
cluster_assignments.png	DAPC cluster probability plot
tree_plot1/2.png	Annotated phylogenetic tree
map_all_years.png, map_2009_2011.png	Spatial maps of reproduction mode
effect_of_latitude_*.png	Regression plots

---